An Exploration of Online Consumer Ratings Systems

Online consumer ratings systems are an important source of information for consumers as they make purchasing decisions. In this dissertation, my coauthors and I elucidate how the way that consumers are asked to provide ratings affects the ratings they give and how different aspects of an existing ratings distribution affects consumers’ product impressions. In Chapter 1, we show that when consumers are asked to rate the overall quality and several attributes of a subpar experience (e.g., the food, service, ambiance, and value at a restaurant), they give a significantly higher overall rating compared to when they are only asked to rate the overall experience. Consumers give higher overall ratings when they can directly rate what went wrong with the experience via the attribute ratings. In Chapter 2, we show that consumers prefer products with rater disagreement (e.g., one rater gives 1 star and one rater gives 5 stars) over products with rater agreement (e.g., both raters give 3 stars) when there are only a few ratings. We show that this effect flips when there are many ratings and explain these findings with our novel theoretical framework, the diagnosticity account.